Re, Makes sense to read the documentation, I will try with another preconditioners.
Thanks for the support. 2016-06-02 18:15 GMT+02:00 Matthew Knepley <[email protected]>: > On Thu, Jun 2, 2016 at 11:10 AM, neok m4700 <[email protected]> wrote: > >> Hi Satish, >> >> Thanks for the correction. >> >> The error message is now slightly different, but the result is the same >> (serial runs fine, parallel with mpirun fails with following error): >> > > Now the error is correct. You are asking to run ICC in parallel, which we > do not support. It is telling you > to look at the table of available solvers. > > Thanks, > > Matt > > >> [0] KSPSolve() line 599 in <...>/src/ksp/ksp/interface/itfunc.c >> [0] KSPSetUp() line 390 in <...>/src/ksp/ksp/interface/itfunc.c >> [0] PCSetUp() line 968 in <...>/src/ksp/pc/interface/precon.c >> [0] PCSetUp_ICC() line 21 in<...>/src/ksp/pc/impls/factor/icc/icc.c >> [0] MatGetFactor() line 4291 in <...>/src/mat/interface/matrix.c >> [0] See http://www.mcs.anl.gov/petsc/documentation/linearsolvertable.html >> for possible LU and Cholesky solvers >> [0] Could not locate a solver package. Perhaps you must ./configure with >> --download-<package> >> >> >> >> >> 2016-06-02 17:58 GMT+02:00 Satish Balay <[email protected]>: >> >>> with petsc-master - you would have to use petsc4py-master. >>> >>> i.e try petsc-eab7b92 with petsc4py-6e8e093 >>> >>> Satish >>> >>> >>> On Thu, 2 Jun 2016, neok m4700 wrote: >>> >>> > Hi Matthew, >>> > >>> > I've rebuilt petsc // petsc4py with following versions: >>> > >>> > 3.7.0 // 3.7.0 => same runtime error >>> > 00c67f3 // 3.7.1 => fails to build petsc4py (error below) >>> > 00c67f3 // 6e8e093 => same as above >>> > f1b0812 (latest commit) // 6e8e093 (latest commit) => same as above >>> > >>> > In file included from src/PETSc.c:3:0: >>> > src/petsc4py.PETSc.c: In function >>> > ‘__pyx_pf_8petsc4py_5PETSc_6DMPlex_4createBoxMesh’: >>> > src/petsc4py.PETSc.c:214629:112: error: incompatible type for argument >>> 4 of >>> > ‘DMPlexCreateBoxMesh’ >>> > __pyx_t_4 = >>> > __pyx_f_8petsc4py_5PETSc_CHKERR(DMPlexCreateBoxMesh(__pyx_v_ccomm, >>> > __pyx_v_cdim, __pyx_v_interp, (&__pyx_v_newdm))); if >>> (unlikely(__pyx_t_4 == >>> > -1)) __PYX_ERR(42, 49, __pyx_L1_error) >>> > >>> > using >>> > - numpy 1.11.0 >>> > - openblas 0.2.18 >>> > - openmpi 1.10.2 >>> > >>> > Thanks >>> > >>> > 2016-06-02 16:39 GMT+02:00 Matthew Knepley <[email protected]>: >>> > >>> > > On Thu, Jun 2, 2016 at 9:12 AM, neok m4700 <[email protected]> >>> wrote: >>> > > >>> > >> Hi, >>> > >> >>> > >> I built petsc 3.7.1 and petsc4py 3.7.0 (with openmpi 1.10.2) and >>> ran the >>> > >> examples in the demo directory. >>> > >> >>> > > >>> > > I believe this was fixed in 'master': >>> > > >>> https://bitbucket.org/petsc/petsc/commits/00c67f3b09c0bcda06af5ed306d845d9138e5003 >>> > > >>> > > Is it possible to try this? >>> > > >>> > > Thanks, >>> > > >>> > > Matt >>> > > >>> > > >>> > >> $ python test_mat_ksp.py >>> > >> => runs as expected (serial) >>> > >> >>> > >> $ mpiexec -np 2 python test_mat_ksp.py >>> > >> => fails with the following output: >>> > >> >>> > >> Traceback (most recent call last): >>> > >> File "<...>/demo/kspsolve/test_mat_ksp.py", line 15, in <module> >>> > >> execfile('petsc-ksp.py') >>> > >> File "<...>/demo/kspsolve/test_mat_ksp.py", line 6, in execfile >>> > >> try: exec(fh.read()+"\n", globals, locals) >>> > >> File "<string>", line 15, in <module> >>> > >> File "PETSc/KSP.pyx", line 384, in petsc4py.PETSc.KSP.solve >>> > >> (src/petsc4py.PETSc.c:153555) >>> > >> petsc4py.PETSc.Error: error code 92 >>> > >> [0] KSPSolve() line 599 in <...>/src/ksp/ksp/interface/itfunc.c >>> > >> [0] KSPSetUp() line 390 in <...>/src/ksp/ksp/interface/itfunc.c >>> > >> [0] PCSetUp() line 968 in <...>/src/ksp/pc/interface/precon.c >>> > >> [0] PCSetUp_ICC() line 21 in <...>/src/ksp/pc/impls/factor/icc/icc.c >>> > >> [0] MatGetFactor() line 4240 in<...>/src/mat/interface/matrix.c >>> > >> [0] You cannot overwrite this option since that will conflict with >>> other >>> > >> previously set options >>> > >> [0] Could not locate solver package (null). Perhaps you must >>> ./configure >>> > >> with --download-(null) >>> > >> >>> > >> <...> >>> > >> ------------------------------------------------------- >>> > >> Primary job terminated normally, but 1 process returned >>> > >> a non-zero exit code.. Per user-direction, the job has been aborted. >>> > >> ------------------------------------------------------- >>> > >> >>> -------------------------------------------------------------------------- >>> > >> mpirun detected that one or more processes exited with non-zero >>> status, >>> > >> thus causing >>> > >> the job to be terminated. The first process to do so was: >>> > >> >>> > >> Process name: [[23110,1],0] >>> > >> Exit code: 1 >>> > >> >>> -------------------------------------------------------------------------- >>> > >> >>> > >> >>> > >> What have I done wrong ? >>> > >> >>> > >> >>> > >> >>> > >> >>> > > >>> > > >>> > > -- >>> > > What most experimenters take for granted before they begin their >>> > > experiments is infinitely more interesting than any results to which >>> their >>> > > experiments lead. >>> > > -- Norbert Wiener >>> > > >>> > >>> >> >> > > > -- > What most experimenters take for granted before they begin their > experiments is infinitely more interesting than any results to which their > experiments lead. > -- Norbert Wiener >
